The purpose of this project is to develop the methodology of cancer control epidemiology, with emphasis on its analytic models, its data tracking needs, and its foundations in biological and statistical theory. Research areas include: concepts of cause and prevention; logic and methods of inference; alternative models to cause and chance; the scientific, engineering, technological and ethical implications of cancer control research; the relationship of carcinogesis models to models of interaction; problems of self-selection and selection bias; using absolute and relative measures to assess the effects of cancer control interventions; and an evaluation of the data-tracking needs of large defined populations suitable for cancer control intervention studies.